The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation
Paper • 2609.36484 • Published • 565
Megan Garcia
megangar
·
AI & ML interests
natural language understanding, reinforcement learning, mechanistic interpretability, question answering
Recent Activity
liked a model about 5 hours ago
autotrust/JEV-35B updated a model 4 days ago
megangar/clip-checkpoint published a model 4 days ago
megangar/clip-checkpointOrganizations
None yet
upvoted 2
papers 18 days ago